The long term objectives of this project are to: (a) develop antiretroviral treatment strategies that minimize the emergence of drug resistance particularly to the most clinically promising combination regimens and (b) gain insight into reverse transcriptase (RT) function by studying resistance mechanisms. The studies will focus on the following specific aims: 1. To perform surveillance for multi-drug resistance development in vivo. Isolates from subjects treated with combination regimens will be studied by standardized susceptibility testing on peripheral blood mononuclear cells (PBMC) and DNA sequencing. We will ascertain if isolates contain an accumulation of RT mutations seen with monotherapy or if novel mutations occur under combination regimen selection pressure. Multi-drug resistance (MDR) is defined as resistance to each of the multiple agents in a combination regimen encoded within a single HIV-1 genome. An assay, utilizing cloned RNA PCR products to reconstruct a deleted provirus to contain RT sequences from circulating blood, will define whether MDR occurs in vivo. Sensitivity of multiple virus clones to different drugs will be assayed by plaque reduction in CD4-positive HeLa cells and each clone will be sequenced. By correlating these results with those of the virus pool, we will evaluate if mixtures of singly resistant viruses can lead to multiply resistant isolates. A different MDR screening test, applicable to both RT inhibitors and agents directed against other HIV-1 targets, will also be developed to assess drug susceptibility phenotype in PBMC. Drug resistance will also be used to determine whether many HIV-l proviruses remain latent in PBMC. 2. To select for HIV-l resistance by serial cell culture passage. We will evaluate cumulative development of resistance on different genetic backgrounds and determine whether resistance emerges more rapidly in syncytium-inducing (SI) viruses than non-syncytium-inducing (NSI) viruses. 3. To study the enzymology of RT in vitro. We will define the mechanisms of resistance to zidovudine and other nucleosides by studying enzyme- template interactions, including polymerase processivity. We will use RNAs to inhibit template strand transfer by the RT and explore if this can prevent retroviral replication. Testing for dideoxyadenosine triphosphate (ddATP) susceptibility of polyethylene glycol (PEG)-precipitated virion RT will be compared to standardized virus didanosine (ddI) susceptibility testing. The virion RT assay is more rapid than virus assays now in use. We will continue to characterize combinations of drug resistance mutations that severely impair enzymatic function using a rapid, site-directed method for mutagenizing RT expression vectors.